PME Treasure

Run an external Python script from PME

2 min read

Outcome

Put longer Python in a readable file, then call it from a PME item with execute_script().

This keeps multi-step logic out of a long Command one-liner. The file can be formatted, reviewed, reused, and tested independently.

Prerequisites

  • PME 2.1 installed and enabled.
  • A script whose source and effects you have checked.
  • The PME user resource folder initialized.

Steps

  1. Open the PME user resource folder. Its default Windows path is:

    %USERPROFILE%\Documents\Pie Menu Editor\

  2. Create scripts/hello_world.py with this content:

    message_box(kwargs.get("message", "Hello from PME"))
return_value = True

  3. Create or edit a PME item and open its Command tab.

  4. Enter:

    execute_script("scripts/hello_world.py", message="External script is working")

  5. Apply the item and invoke it.

PME resolves scripts/... against the user’s scripts directory first, then its bundled system scripts. An absolute path is also accepted, but a resource-relative path is more portable.

Data available inside the script

The script receives:

  • kwargs — values passed by the calling item;
  • __file__ — the resolved script path;
  • return_value — set this when the caller needs a result;
  • PME globals such as C, D, O, L, and the current scoped event when one exists.

Verify

The message box should display External script is working. If it does not:

  1. confirm the file is inside the active resource root’s scripts folder;
  2. check the spelling and filename extension;
  3. open Blender’s System Console for the Python exception;
  4. reduce the script to the minimal example above before debugging its real logic.

Limits and safety

Python scripts can modify Blender data, preferences, and files on disk. Review unfamiliar code before running it, and test destructive or batch operations in a copy of the .blend file.

The 2019 forum discussion includes a multi-object modifier example. Its object-mode assumptions and Blender API calls need fresh validation before reuse, so this guide keeps the example focused on execute_script() itself.

Sources

Graph View

Backlinks